Econometrics: one step ahead forecast - page 2

New comment
 
faa1947:
Don't read Soviet newspapers!
Well, yes, the news is less believable than the leaking model on the spread on the chondrik.
 
faa1947:

This has nothing to do with Hodrikt.

The model used has the idea of separating out the deterministic component and adding noise to it .

There are other ideas. What about you? Is there gas in your flat? Or do you have ideas, please, and I'll show you the calculations.


well... go ahead ;))

There are plenty of ideas. But you should try to generate your own ideas, instead of being just a calculator, it would be more useful.

I can calculate my own ideas.

 

Taking the next step.

I make a new opening price = 1.3524
.

Since the model is junk, in accordance with the wishes above, let's increase the number of lags and take the following model:

kotir hp1(-1 to -4) hp1_d(-1 to -4)

Let's estimate this regression:

Well, much better, the R-squared is very decent, but the regression error is very large (S.E. = 55 pips), most importantly, there are two coefficients with a high probability of being equal to zero. Let's remove two lags to model the error and estimate y-e:

kotir hp1(-1 to -4) hp1_d(-1 to -2)

Result:

Great, let's calculate the result:

Forecasting a long for today.

But the results along the sample are dismal.


 
avtomat:

Well... go ahead ;))

There are plenty of ideas. But you should try to generate your own ideas instead of being just a calculator - it would be more useful.

I can calculate my own ideas.

What are you doing on the forum?
 
TheXpert:
Well yes, the news is less believable than the leaking model on the spread on the chondrick.
To each his own. I'm trying to get a calculation. Yesterday's prediction was correct.
 
faa1947:
What are you doing on the forum?

guessing ;)))))))
 
faa1947:

The result of the previous forecast.

Forecast was for a short - have a short - forecast successful!



"So: Forecast for tomorrow from 00:00 hours 1.3798."

And there's a fake one! (mana on ? - mana on mana)

 
avtomat:
guessing ;)))))))
sympathy ;)))))
 
Tantrik:

"So tomorrow's forecast from 0000 hours 1.3798."

And there's a fudge! (mana na ? - mana na mana)

I don't believe in manipulation. Why would he do that? It's a study, not a measurement of predictions.
But you have to put a number on it to make it right.
 
Tantrik:

"So tomorrow's forecast from 0000 hours 1.3798."

And there's a fudge! (mana na ? - mana na mana)

And it's rigged!

Careful what you say.

We are talking about the current 24 hours from 00:00


123456789...139
New comment